Markets, Systems& Time Frames
The first step in developing a trading system is to select the market action and corresponding system type that you want to trade. As I've discussed, selecting a system type is a very important part of system trading and you should take your time in evaluating the alternatives. Many factors will influence your decision, but your own personality will ultimately direct you to the system that is right for you. In making the choice, the most important thing to remember is that it is yours to make alone. Read everything I have to share with you about different types of systems, but then decide for yourself. Only you really know what type of person you are and therefore what type of trading is best for you.
This chapter will help you to understand some of the conditions that can occur in the market, and the system type that complements those conditions. Once you are familiar with the basic system types, you will be able to select the one you want to use.
Three Market Types
Generally, there are three types of markets. The three market types, or phases, are derived from three distinct chart patterns that appear when there is 858x2315i a shift in market action. The phases are trending, volatile, and directionless, and each can be characterized by specific price activity. Take a look at the following charts and familiarize yourself with each different market pattern.
TRENDING MARKET
A sustained large increase or decrease in price characterizes a trending market. Take a look at Chart 1. This weekly chart of Coca Cola ( ) from early to mid-1997:
In fact, this stock has been in an up-trend since 1994. has almost tripled since then. This trending market was characterized by sustained up moves with very small and short-lived corrections. The 9- and the 18-period moving averages are included in Chart 1. A trend trader would buy the market when the shorter 9-period moving average crosses above the 18, and hold the stock until the 9-period average crosses below the 18. In this time period, he would have held for at least two trend moves.
Now take a look at this daily chart, Chart 2, of the Swiss Franc from mid-1996 to early 1997:
In this time period, the Swiss Franc has been in a daily downtrend for many months. It has lost more than 15% of its value over the period. This market was characterized by a sustained downmove with very small corrections. The same moving averages were plotted here, the 9 and 18. Note that if you had followed these averages, you would have stayed short for several months at a time.
The time frame you are looking at is important when you consider the type market action. Chart 3 shows the same Swiss Franc viewed on a monthly instead of daily chart.
The downtrend in 1996-1997 looks a little different when put in this perspective. It looks like the most recent move in a directionless market. And if you had traded the same moving averages on Chart 3, you would have been chopped around and most likely lost a lot of money. The point is that you should be aware that a directionless monthly or weekly chart might have very tradable daily trends, and vice versa.
DIRECTIONLESS MARKET
A directionless market is characterized by smaller, insignificant up and down movements in price, with the general movement sideways. We probably would not call Chart 3 of the Swiss Franc directionless because the movements were not insignificant.
On the other hand. Chart 4 of Caterpillar in 1996 clearly shows a sideways directionless market, whose movements I would call insignificant, as the stock moved between 31 and 37 for most of the year. Markets chop around like this between trends. As you can see, I put the Stochastic Indicator on this chart. The Stochastic Indicator is commonly used as an overbought/oversold indicator. In directionless markets, you might attempt to buy CAT when the Stochastic is at or below 20 or 25 and sell when it is above 75 or 80. You could have made some money doing this with CAT in 1996.
VOLATILE MARKET
A volatile market is characterized by sharp jumps in price. Chart 5 is a weekly chart of American Software. You will notice that this type of market action involves a quick and unexpected change in volatility. At the marked points on this chart, AMSWA was quiet for the previous 7 to 15 weeks. Then the price leaped out of this low volatility trading range. This is what is commonly called a "volatility expansion."
The volatility of the market increased substantially during the breakout week as it shot out of the previous range. Systems can be designed to take significant drawdown. Therefore, if your system misses a big move, you may not have enough capital to hold out through the drawdown for the next big move.
Another design priority should be to limit your losses during the market's sideways mode. Notice how I said limit losses not make profits. It is very important to recognize that no system will make money in every market condition. It is therefore very important to identify the market action in which the system will make money and the market action in which it will lose money.
Once you have found the market action in which the system will lose money, it becomes a system design priority to minimize losses during that market action. If the system is designed to make money in a trending market, it will lose money in the choppy phase. Your priority should be to minimize the losses in the directionless market.
Many trend-following systems make their money in one or two trades of the year and break even or lose money for the rest. The most common indicator used for trend following is moving averages, most often two, a short moving average and a longer moving average. Chart 6 of Disney shows the 9- and 18-period moving averages with TradeStation arrows indicating where a 9- and 18-period moving average crossover system would go long (up arrow) and short (down arrow).
As you can see, there were periods of trend where a significant amount of money was made as well as periods where the market was choppy and the system whipsawed back and forth with losses.
Let's analyze what we've just learned. Most trend traders will tell you that the 80/20 rule works for trend trading: they make 80% of their profits on 20% of their trades. Even though the moving average system on Disney (Chart 6) made money over time, it was profitable only 39% of the time. That means that the system lost money 61% of the time. This is the difficult part of trend trading-a low percentage of winning trades. You need a lot of positive self-esteem and a lot of confidence in your abilities to trade a system that loses money on 60 or 65% of its trades.
We will talk about this issue again later, but you should be thinking now about the design of the system you would be able to trade. If you want to be a typical trend trader, you should be prepared to lose money in a majority of trades. You should also be able to sit through significant drawdown as the market drifts through a directionless period.
The table below, SPF 1, is what I call a System Parameter File. It is a summary of all the relevant information that I use to create a system with TradeStation. Each time I test a system in this book, I will use this so that you can see a description of a system in summary form and you have all the information to reproduce the results if you so desire.
Look at the Performance Summary labeled PS 1. As I just asked you, could you sit in front of your computer screen and place losing trade after losing trade, waiting for the big move to come? Could you sit through a string of 6 or 7 losses in a row before the next profitable trade? Could you lose $20 per share in a string of losses?
Note: If you are unfamiliar with Performance Summaries (System Report), please refer to Chapter 8, The Science of System Evaluation.
As you can see from PS 1, the maximum number of consecutive losers was 6 and the maximum infra-day drawdown (MAXID) was $20.13. That means that at least once, from 1990 to 1997, you would have placed six losing trades in a row and had a cumulative loss of over $20 per share. Could you realistically put up with this?
Another characteristic of a trend-following system is that it makes most of its profits in one or two big trades. Of the $18 profit in Disney, $10 came from one trade over the six years of data. This is not unusual for a trend-following system. I discuss how much profit you should permit to come from the largest profitable trade in Chapter 8, The Science of System Evaluation.
Many researchers have estimated that any market is in the trend mode 15% of the time and is directionless 85% of the time. A trend-following system then, by definition, has a low percentage of profitable trades. A trend-following system is psychologically difficult to trade, but if you think you can successfully trade without constant positive feedback, it can prove to be very profitable.
Trend-following systems are probably the most popular type of system. With a high percentage of losing trades, you might be wondering why is it so popular. Very simply, trend-following systems can be very profitable over time. Another reason is that people like to follow (and make money on) the big trends. It is human nature to want to cash in on the big moves in the market. It is innately satisfying to get in early on a trend and watch your profits soar.
SUPPORT & RESISTANCE SYSTEMS
The main focus of a Support and Resistance (S/R) system is to profit from the price swings that occur in directionless markets. The system attempts to capture price movement opposite to that captured by trend-following systems.
Support and resistance systems start with the premise that markets are directionless 85% of the time. The system attempts to take advantage of this price movement and catch the small swings that take place in sideways or choppy markets.
This type of system has a higher number of winning trades, with small profits on each trade. It misses the full trend because it exits early in the trend move as the market becomes quickly overbought or oversold.
An S/R system is built on the concept of buying low and selling high. As you are buying when prices are low and selling when prices go up, you are actually trading against the trend. Essentially, you are attempting to pick tops and bottoms. You buy low and sell high, but the market keeps going higher. You keep selling as the market goes higher, and keep taking small losses until the market finally turns down and gives you a profitable trade.
Although an S/R system is easier to trade emotionally, many traders don't trade this type of system because they miss the big move (by design). The most common indicator used with a support/resistance type of system is probably the Stochastic Oscillator.
You can see the Stochastic Indicator on Chart 7 of Caterpillar. I also applied the Stochastic Crossover system I created based on this indicator, highlighted in SPF 2.
Notice how the indicator fluctuates between 0 and 100. In this case, I used the 65 line and the 35 line to represent overbought and oversold, respectively. The overbought level for the stochastic is generally between 65 and 90 and the oversold level is between 35 and 10. You can play around with these levels to find the ones that make the most sense for you.
I have designed an S/R system so that when the Stochastic (SlowD) is below 35 and the short average (SlowK.) moves above the long average (SlowD), the system produces a buy signal. The opposite would be true for a short signal, SlowD is above 65 and SlowK crosses below SlowD.
The drawback of support and resistance systems is that they usually have small profits and larger strings of losses as they lose money when the market trends. By design, the system keeps shorting a market that is in an uptrend, or buying a market that is a downtrend.
You can see this happened twice in Chart 7 (previous page). Both times the market was in a sustained up-trend and when the Stochastic set-up reached overbought (above 65), the systems went short. The market then kept moving up, resulting in losing trades.
As you can see from Performance Summary PS 2, this system has a high percentage of profitable trades (68%). This high percentage is needed to be profitable overall because the average losing trade was close to 1/3 larger than the average winning trade. Observe also that the system only had two consecutive losses in a row, which makes it much easier to trade from a self-esteem standpoint. The maximum intra-day drawdown (MAXID) was very large as a percentage of the net profit (76%). This would have to be fixed before this system would be ready to trade. I'll show you techniques for fixing problems like this in the following chapters.
Keep in mind that, while system development looks easy, it is not. CAT was in a choppy market during this time whereas the stock market was in a strong bull market. If you had traded our moving average crossover trend-following system on CAT during this time, you would have lost a significant amount of money thinking that CAT would trend with the overall market.
An S/R system is designed to buy low and sell high, which is an easy method psychologically to trade because it makes logical sense. However, these systems can lose money in the long run. Generally, most successful system traders don't trade this type of system. If S/R systems are used at all, it is to complement a group of systems that includes trending systems and perhaps a volatility system or two.
VOLATILITY EXPANSION SYSTEM
Volatility expansion systems are designed to do well in volatile markets. The trades generated by this type of system are usually short-term, and when trading this type of system, you will be out of the market a significant amount of time. Volatility expansion systems generate a high percentage of winning trades, although these trades usually generate small profits per trade. The S&P futures is a market that I would characterize as "volatile." Neither trend-following systems nor S/R systems work particularly well on the S&P.
Chart 8 is a daily S&P futures chart from December of 1996 through March of 1997. Using a ShowMe Study, I had TradeStation highlight the gaps by placing large crosses on the opening price on the day on which the gap occurred.
One characteristic of a volatile market is gaps. Gaps refer to places in a bar chart where there is no continuity or overlapping of price. In this case, I have defined a gap as existing when today's open is either above the high of yesterday or below yesterday's low. Chart 8A is a small chart with two examples of up gaps.
|
|
As you can see from the marks on the S&P in dart 8 (previous page), gaps appear to indicate that the market makes substantial daily moves following an opening gap. Let's try to capture this movement with a system that is designed to profit from opening gaps and subsequent movement.
Let's assume that if the market gaps up it is going to continue to go up, and it 11 gaps down it is going to continue to go down. The up or down gap sets up the trade. We then need to figure out how we are going to enter the market once the set-up occurs. I think we should require that the market move a significant amount away from the opening price before we enter the market.
This system enters when the price action moves up or down an amount equal to yesterday's close plus or minus yesterday's range. The idea is that in addition to the price gap on the opening, we will require the price to move a distance at least equal to the previous day's range away from the previous day's close. This adds a second condition, assuring that volatility actually does expand. The system is applied to a daily S&P futures chart in Chart 9.
The results of this system are pretty good for the first try. This is definitely something that we can work with. There are many additions and variations that could improve the system. We might work on different exits, money management stops, and profit targets. We might also work on different ways of entering the market after a gap occurs. The results in PS 3 indicate that this price movement has real potential for creating a viable system.
As you can see in PS 3, the profits from a volatility expansion system come from a high percentage of profitable trades. Even though the average winning trade was less than the average losing trade, we still had the makings of a profitable system.
I hope you also noticed that in this test, as in all the previous tests in this chapter, I did not include any costs for slippage and commission. If, for instance, we included $25 for commission and $75 for slippage, the average trade profit would be $91.13 instead of $191.13. In systems that have a lot of trades, these costs can make the difference between a system you would trade and one you would not.
Price explosions of one form or another characterize a volatile market. One way of defining a price explosion would be a "gap" opening, another would be an increase of "range" (high-low). Some indicators have been developed to try to indicate a change in volatility. One of these is actually called "volatility" and is included as a study in TradeStation.
Typical volatility expansion systems measure current volatility and enter the market when there is an abrupt increase in volatility. This type of system makes a quick exit, usually after only a few bars.
Selecting a Market and System Type
You should now have an idea as to the different types of market action and the system characteristics that attempt to take advantage of the action and profit from it. Each type of market has unique characteristics and takes a different thought process for system design.
In your own thoughts, you should begin to think about what type of market you are most comfortable with and would like to trade. Another consideration is the financial and statistical characteristics of the systems, with specific regard as to whether you could actually trade the system. It is not wise to create a great system that would be psychologically impossible for you to trade.
The first step in system design is to think about the characteristics of the three market types and the systems that are effective for each. Then decide what type of trader you are, or want to be: a trend trader, who buys low and sells high, or a volatility trader, who takes selective but high percentage trades.
I don't want to tell you what kind of system you should use. Everyone has to decide for him or herself, based on their personality and trading preferences. I think the best way to choose a system is to take a look at Table 1.
You should determine what type of system is best for your temperament. There are successful system traders using each type of system, but based on my experience, a higher number of traders use trend-following and volatility expansion systems than support and resistance systems.
Choosing a Time Frame
After you select the system type you want to use, you need to think about the rime frame in which you want to trade, and therefore the type of data you want to collect. There are three general types of data you can collect: intra-day, daily, or weekly. Choosing the time frame that is appropriate for you is almost as important as the type of market action and system you want to trade.
The most common chart used by traders is the daily chart, and this is why I use daily charts for most of the examples in this book. Daily charts are the most common for several reasons. Because most traders also have day jobs, they want to keep abreast of the market as much as possible without it intruding into their workday. The daily chart is perfect for this type of trader. You are able to review the markets each night and make your decisions for the next day.
WEEKLY VS. DAILY CHARTS
Weekly charts are much more difficult to trade because it takes more discipline. To trade weekly charts, you must make your decisions on the weekends and not make any changes until the next weekend. For most traders, this is very difficult to do. It is very easy to yield to temptation and move a stop loss or a money management stop, or want to keep your profits and exit the market early.
To discipline yourself not to look at the market during the week is a tough thing to do. Most people don't think of trading weekly charts. My experience is that there is a lot of money to be made trading weekly charts, simply because so few traders are able to do so. To make money in the markets, you have to tread where the average traders do not tread. Weekly charts are one of those places.
Chart 10 shows the weekly S&P futures in the upper box and the daily S&P in the lower.
There is more price detail in the daily chart, but also more price noise. Let's check out a simple system on both the daily and weekly charts.
Chart 11 is a daily IBM with a 50-period channel. The 50-period channel marks the highest high of the last 50 bars and the lowest low of the last 50 bars. The system would go long if the close of the bar closed above the channel and sell short if the price closed below the lowest low of the last 50 bars.
Chart 11 shows IBM up to and including the crash of 1987. The first thing you should notice is that this system is always in the market, i.e., it is either long or short. I arbitrarily chose the 50-period channel for this test. I will then compare the results with the same length channel on a weekly chart.
In these tests, I will assume that 50 days is about equal to 10 weeks. So, to compare a daily system with a weekly system, we will use the same lengths in time although measuring the length on daily charts in days (50) and on weekly charts in weeks (10).
Let's take a look at how a simple channel breakout system works, first on a daily chart, then on a weekly. Our working premise is that the system will be more profitable on weekly charts than on daily.
Ask yourself why should a system, basically the same system, work better on a weekly chart than on a daily. I can come up with several reasons. First, very few people have the patience and the discipline to trade weekly charts. Second, by their very nature weekly charts smooth the price fluctuations of the daily chart. If there is a long trending market, we should be in the trend
longer. We might get in the trend a little later than on the daily chart, and out later, but we will probably not get whipsawed as much in the directionless markets.
I chose IBM again arbitrarily because it went through some frustrating choppy periods and some very fine trending periods in its action packed history since 1970. The System Parameter File SPF 4 shows how we would design a system to test this theory.
Let's look at the results for the daily chart, shown in PS 4.
This system was profitable over the 27 years. IBM moved from a low of 24 to a high of approximately 110, an 86 point rise. The system made $29 per share from 1970 to 1997.
Now let's look at the same indicator and system on an IBM weekly chart, Chart 12.
Keep in mind that this is essentially the same indicator and system as the daily chart. The System Parameter File is shown in SPF 5. Notice that the only difference is that it is a 10-period channel on a weekly chart instead of a 50-period channel on a daily chart.
All else being equal, the systems should perform about the same. However, as you can see in PS 5, in almost every category the weekly system outperformed the daily system.
Both systems took their first trade within two days of April 22, 1970. From that point on, the weekly chart had a greater profit on fewer trades and less drawdown than the daily chart. The rest of the data is about the same. Clearly this data comes down on the side of the weekly chart rather than the daily.
This is just one very simple example of why you should consider weekly charts and not just assume that daily charts are your only option for trading.
INTRA-DAY VS. DAILY CHARTS
Intra-day charts are the 5-, 10-, 30-, and 60-minute charts that are compiled from intra-day tick data. To trade intra-day charts, you must give almost your full attention to the markets during the day.
It is virtually impossible to have a full-time job and trade intra-day charts well. As a percentage of traders, relatively few traders are able to trade during the day. I think it is for this reason that there is significant money to be made trading intra-day. The relative lack of competition has to be in your favor trading intra-day. Chart 13 is an example of a 30-minute S&P futures chart placed on top of a daily chart.
Trading intra-day data permits you to put a microscope on daily activity and filter trades so that you can take advantage of the intra-day timing. I want to show you the benefits of looking at a technique and system through the intra-day microscope.
To do so, let's analyze a technique that I taught in my seminars many years ago. 1 called it a RangeLeader Breakout. A range leader is a special type of bar that has two attributes. The first is that the range of the bar must be greater than the range of the previous bar. Range is defined as the bar's high minus the bar's low.
The second characteristic of a range leader is that the midpoint of the bar must be above the previous bar's high or below the previous bar's low.
So let's create a system using the range leader. And make it simple. If a range leader occurs today, on the current bar, we will buy tomorrow one tick over the high of the range leader, or we will sell one tick below the low of the range leader. That's about as simple as I can conceive it.
The daily chart of the S&P with both the ShowMe Study and the RangeLeader Breakout system on TradeStation is shown in Chart 14.
What type of a system is this? Trend-following, support and resistance, or volatility expansion? This the first question you should ask yourself as you look at this or any other system. In this case, since we're looking at a breakout based on the previous bar's range, it is a volatility expansion system.
The System Parameter File is shown in SPF 6.
Note that this system introduces the concept of Money Management Stops ("MMS") and Price Targets ("PT"). An MMS is an order you place in the market to conserve your capital. In this case I decided I did not want to risk more than $500 per trade. The system design therefore includes a provision that when it gets filled, it immediately puts a stop loss $500 away from the entry price.
Price targets are placed if you want to exit the market at a particular profit level. This, of course, limits your profit per trade. In this case, I decided that if the price moved $1, 500 in my favor, I would take the profit. For this system, I arbitrarily decided on the $500 MMS and $1, 500 PT amount, but if we want to we can use TradeStation to test for the optimum amount for both of these.
For the exit, if my price target was not hit, that is, I did not make the $1, 500, 1 decided that I would want to get out a soon as possible. I had two choices as to how to exit: on the close of the entry bar or the next day on the open. I chose the next day on the open as I wanted to take advantage of possible gap opens. I could also test other options for exiting the market. So, if I did not make $1, 500 on the day the system entered the market, I would exit the following day on the open. This system was not too bad right out of the box.
The Performance Summary for this system is shown in PS 6.
We made 125% return on our drawdown in one year. All in all, not a bad first try.
Are we able to improve on this basic system by using the microscope of infra-day charts? Let's try using a 30-minute chart and see what we find. The Performance Summary results are shown in PS 7.
This obviously didn't work. We simply put the same system for the daily chart on the 30-minute chart (with one small change). The System Parameter File is shown in SPF 7.
Again, the system entered on RangeLeader Breakouts with a $1, 500 PT and a $500 MMS. This time it lost some money. Let's look at the minor change I made to it and then think for a moment about what went wrong.
An important consideration for this system, as with any intra-day system, is the first and last bar of the day. If the first bar of the day is a range leader, this means that the range of this bar is greater than the range of yesterday's last 30-minute bar, and that the mid-point of this bar is either greater than the high or less than the low of the last bar yesterday. I have always thought that with the intervening time, this information was meaningless and shouldn't be used to trade.
Therefore, I added a Condition 1, which eliminates the first bar from use (SesslStartTime). Notice that I have also eliminated the last bar of the day (SesslEndTime). If the last bar of the day is a range leader, the breakout will occur tomorrow during the opening bar. The fact that the last bar of the day is a range leader is irrelevant to tomorrow's first bar, and the breakout is meaningless. Condition 1 and Condition2 in the TradeStation EasyLanguage for this system deal with these issues.
Now, let's look now at what went wrong. I believe the reason that the system lost money on the intra-day chart is because we didn't take advantage of the strengths of using intra-day charts. It stands to reason that there must be certain times of the day when the market moves and other times when it rests. We simply used each 30-minute bar as if it was no different than any other bar. I have always thought that there were different times of the day that are more important.
Perhaps we should test each individual bar for a RangeLeader Breakout and then put in our MMS and PT and exit on tomorrow's opening price if we don't reach our target or get stopped out.
There are 13 30-minute bars during the day, and a 14th bar which is the last 15 minutes between 4: 00 and 4: 15pm (EST). I changed the system to test each bar, designated by its ending time, for a RangeLeader Breakout, using a $1, 500 PT and a $500 MMS. If neither the MMS nor the PT is hit, we then exit the next day on the open. A summary of the results for each 30-minute intra-day bar is shown in PS 8.
PS 8 shows that there were only three time periods that produced profitable trades, 11: 00, 12: 30 and 15: 00 (3: 00pm). Clearly the 15: 00 bar was the most profitable. It looks like we can conclude that most of the action in the S&P takes place after 3: 00 in the afternoon.
SPF 8 shows a summary of the final design of the 3: 00 intra-day RangeLeader Breakout. Key elements of this system are the time of day, the $1, 500 profit target, the $500 money management stop, and the exit on the open of the following day if neither of the stops are hit. PS 8A shows the whole Performance Summary using only the 15: 00 bar as the RangeLeader Breakout.
As you can see in PS 8A, the results of using 30-minute bars and only using the 15: 00 bar the RangeLeader Breakout system were very good. In 1993, it produced a return on maximum intra-day drawdown (MAXID) of 659% with 57% profitable trades. In every category, this system outperformed the daily chart.
Keep in mind that this is only for 1993. Before I would get overly excited about this system, I would test this in other years as well.
So, after all of this information, what's the point? The point is that intra-day data, if used correctly, can give you a distinct advantage over daily charts. If you have the time and energy, you can take advantage of the microscopic look at the markets using intra-day charts, and you may be able to improve your return.
Summary
Let's recap what we have covered in this chapter. First, we took a look at the three types of markets: trending, directionless and volatile. We noted their individual characteristics and how to recognize each of them.
Next we studied systems that take advantage of the three different types of market action. First, we looked at trending markets and the trend-following systems that attempt to profit from this type of market. We saw that this type of system tries to catch the big move, and usually loses money while it waits for the trend. Trend-following systems take trades with a low probability of profit, with the eventual profitable trade usually being a big winner, covering all of the losses and more.
We then looked at both support and resistance systems and volatility expansion systems and noted their characteristics. Generally, these systems are designed to intentionally miss the big trend. They attempt to make money by entering trades that have a high probability of success, but have limited profits. S/R systems buy low and sell high. Volatility expansion systems capture an increase in volatility and profit from this short-term explosion in price.
We then looked at the different time frames available for the system trader. I noted that most people instinctively trade daily charts. However, the successful system trader looks at the time frames that will maximize profits, not necessarily those that are most convenient. We compared the same system on the same data on both a daily chart and a weekly chart, and found that in this case the weekly results were much better than the daily. While this won't be the case for every system in every market, it makes the point that using weekly charts is something you should at least consider.
We then turned to intra-day charts. I hoped to show you that the same issue exists for intra-day charts. Are there markets and systems that would be improved by using intra-day charts rather than daily? We found at least one instance where this was true, using my concept of range leaders for an S&P
system.
Our first step was to test an indicator, the RangeLeader, and use it to develop a reasonable system on a daily chart. We then modified the daily system for intra-day data, eliminating the first and last intra-day bar. This didn't work. Undaunted, for the next step we decided to use the 30-minute intra-day data as a microscope to find the periods that did work with intra-day, 30-minute range leaders. For that reason, the last step was to test each of the individual bars to see which bars (if any) produced a viable system. We found the 1, 500 bar to be very profitable and modified our system accordingly.
In this chapter, I hoped to show you that it is not necessary to be locked into trading daily charts. Although daily charts are the most common, and for most people the easiest to use, a case can be made that this is precisely the reason that you should consider trading other time frames. The decision rests on three factors: individual preference, personal discipline, and time.
The move to consider weekly charts involves some self-evaluation. Do you have the discipline to only look at the markets once a week? Can you effectively ignore market action during the week? In many markets, trading weekly charts can be a big advantage; weekly charts tend to smooth out the price action, reducing many of the daily whipsaws into small insignificant corrections. This can be a distinct advantage for trend traders. I showed you one instance where using a weekly chart for a trend system was an advantage.
The intra-day time frame has its own advantages and disadvantages. First, you must have the time to watch the markets during the day. Second, you will probably be entering many more trades, and the cost of commissions becomes a larger factor. And third, the software and data costs are greater.
These are the first issues that you must consider as you begin to develop a system and trade it: the type of market, the system type and the time frame. Let's now move on to the major elements of creating the system itself.
|
